AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers

Zeyuan Allen-Zhu, Yuanzhi Li, Yingyu Liang

NeurIPS 2019 · Architectures and efficiency › RNNs and normalization · approach: Representation geometry and collapse · 154 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Train an overparameterized two-layer ReLU network with SGD on its hidden weightsTwo-layer ReLU learner · Agnostic PAC learning of two-layer smooth-activation target functionspopulation riskImproved
Train an overparameterized three-layer ReLU network with SGD variants that allow hidden-layer interactionsThree-layer ReLU learner · Agnostic PAC learning of three-layer smooth-activation target functionspopulation riskImproved
Use overparameterization so the learner can represent larger target functions and more complex activationsSGD-trained neural networks · Generalization to the population distributionpopulation riskImproved
Use weight decay to scale down the three-layer network during trainingThree-layer ReLU learner · Three-layer learningoptimization stabilityImproved
Use an explicit regularizer that encourages weights to be spread across neuronsThree-layer ReLU learner · Generalization of three-layer networksgeneralizationImproved
Train only the output layer rather than allowing hidden-layer interactionsThree-layer ReLU learner · Learning correlations between nonlinear measurementslearnable function classHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy